Casey Knott
Menu

AI security philosophy

How I think about AI security

These principles come from the projects. They are Casey’s engineering and security philosophy, not a claim that they are universal law.

  1. 01

    AI should not determine its own authority

    Who determines the next action is the load-bearing question. A model can classify, draft, and recommend. It does not grant itself tools, publication rights, or remediation rights.

    Authority is assigned outside the modelLeast privilege / separation of duties

    Grounded in: Multi-Stage Knowledge Publishing Workflow · Knowledge Publishing Agent · Home-Lab Endpoint Triage Agent

  2. 02

    High-impact actions require explicit authorization

    Isolation, process termination, account disablement, publication, customer packaging, and production deploy are privileged. The agent recommends; a person authorizes.

    Human approval gatePrivileged change authorization

    Grounded in: Home-Lab Endpoint Triage Agent · Clearward · Basaltborne · Knowledge Publishing Agent

  3. 03

    Generation and validation are separate concerns

    The same model should not certify its own output. Required fields, hashes, schema, RLS, and style checks belong in deterministic code.

    Deterministic validatorsVerify before trust

    Grounded in: Knowledge Publishing Agent · Clearward · Basaltborne · TinyClaw

  4. 04

    Agents receive only the capabilities they need

    Collection, analysis, and remediation use separate authority where practical. Tool allowlists are the capability boundary. Completing a task more easily with an extra tool is a failed control.

    Tool allowlistAuthorized capability boundary

    Grounded in: Home-Lab Endpoint Triage Agent · Knowledge Publishing Agent · Clearward

  5. 05

    Actions should be observable and auditable

    Log tool calls, inputs, decisions, errors, retries, confidence, approvals, and stopping reasons. If a manipulation attempt is not visible afterward, the control did not hold.

    Audit trailSecurity logging and accountability

    Grounded in: Home-Lab Endpoint Triage Agent · Knowledge Publishing Agent · Basaltborne

  6. 06

    External data is untrusted input

    Logs, documents, websites, filenames, and command output may contain instructions intended to manipulate the model. Tool policy overrides anything asserted in the content.

    Content cannot override policyUntrusted-input handling

    Grounded in: Home-Lab Endpoint Triage Agent · Knowledge Publishing Agent · TinyClaw

  7. 07

    Secrets do not belong in prompts or repositories

    Clearward agent instructions forbid collecting passwords, recovery codes, payment-card data, or MFA secrets, and require a human on CAPTCHA, MFA, or payment. Basaltborne never ships service-role credentials to the browser.

    Human-only credential boundaryCredential protection

    Grounded in: Clearward · Basaltborne

  8. 08

    Failure modes should fail safely

    Stop and escalate when evidence is insufficient, tools fail repeatedly, permissions are exceeded, or confidence requirements are not met. TinyClaw skips a run rather than publish a partial report.

    Safe-stop conditionsFail closed

    Grounded in: Home-Lab Endpoint Triage Agent · TinyClaw · Knowledge Publishing Agent

  9. 09

    Rollback is designed before automation is trusted

    Recommended changes include expected results, validation criteria, operational risk, and a rollback procedure. Lab checkpoints and replayable migrations exist so undo is not improvisation.

    Checkpoints and planned undoRecovery / incident response

    Grounded in: Home-Lab Endpoint Triage Agent · Basaltborne

  10. 10

    Human review belongs where consequence and uncertainty justify it

    Not every lookup needs a committee. Purchases, safety-relevant vehicle changes, publication, customer release, containment, and production deploy do. Review is placed on consequence, not on ritual.

    Review on consequential transitionsRisk-based change control

    Grounded in: Domain-Specific AI Expert · Clearward · Basaltborne · Home-Lab Endpoint Triage Agent